1. Rates and kinds of spontaneous mutations arising in the mesophilic archaeon Haloferax volcanii were determined in a collaboration with Dennis Grogan at the University of Cincinnati. The mutation rate was severalfold lower than predicted and differed when measured in two ways. The mutations were mostly indels of a rare type, additions and deletions of multiples of three base pairs. Then we learned that this archaeon maintains about 20 copies of its chromosome, which will greatly delay the detection of mutations. We suspect that the mutations were mostly strongly dominant alleles. This is an interesting result now submitted for publication, but is not a test of our hypothesis.[unreadable] 2. Rates and kinds of spontaneous mutations arising in the eubacterial thermophile Thermus thermophilus were also determined in collaboration with Dr. Grogan. The mutation rate and fraction of basepair substitutions were indistinguishable from those observed for S. acidocaldarius, in strong support of our hypothesis. These combined results also allow us to reinterpret a classical study to conclude that all three domains of life exhibit the same rate of molecular evolution, but that among the prokaryotes, the thermophiles evolve more slowly.[unreadable] 3. We reported previously that most mutational spectra contain more mutants with multiple mutations than expected from a random distribution of the mutations. The nonrandom distribution of mutations in tobacco mosaic virus has been modeled as the sum of two subpopulations, a majority with a low mutation frequency and a minority with a high mutation frequency. If this result is general, it will help to model the average mutation rate of riboviruses in combination with their rapid evolution within the host but relative genomic stability over time. Now we are establishing a phage system in which such questions can be more efficiently addressed. We also reported that DNA polymerase molecules copying a template in vitro can produce nonrandom bursts of mutations, suggesting that a substantial minority of the polymerase molecules have a mutator phenotype. Now we are establishing systems to determine the molecular basis of such events.